k-Nearest Neighbour (k-NN) classification algorithm has been applied to the dataset in order to classify the songs into albums using computational methods. Unfortunately, k-NN performs poorly on the present dataset. The best obtained score is 5/11 songs correctly classified, that is for Herzeleid. Precision and recall scores, summarized below, indicate an abundance of both false negatives and false positives. For instance, for RAMMSTEIN album, both precision and recall are 0. Low recall means that songs from a particular album are not categorized into the same album by the algorithm. Low precision means that the algorithm also thinks that a lot of songs from other albums belong to an album in question. The only acceptable scores are for Herzeleid, although recall is still quite poor.
| Album | Precision | Recall |
|---|---|---|
| Herzeleid | 0.7142857 | 0.4545455 |
| LIEBE IST FÜR ALLE DA | 0.2500000 | 0.1875000 |
| Mutter | 0.1428571 | 0.1818182 |
| RAMMSTEIN | 0.0000000 | 0.0000000 |
| Reise, Reise | 0.2000000 | 0.1818182 |
| ROSENROT | 0.2307692 | 0.2727273 |
| Sehnsucht | 0.2500000 | 0.2727273 |
Choosing top predictors (loudness, c12, c04) and features used for previous analyses (danceability, energy, valence, loudness and tempo) did not majorly improve things (should I add another table with precisions and recalls for these? The values for both are also awful)
K-NN is a supervised learning algorithm that is useful for classification and regression problems. As was seen from the knn analysis, the performance is poor. So, I have decided to use an unsupervised learning algorithm made to tackle clustering problems, that is k-means. Here, I have tried to group the albums into bigger clusters, which would show which albums are similar and which are different. The distance matrix based on danceability, energy, valence, loudness and tempo shows the distance between album pairs. The closer to 0 (black colour), the less distance, thus the more similar are albums. There is a black diagonal line that indicates that albums are identical, that is distance = 0, because we are comparing an album to itself.
Hierarchical clustering algorithm was used to group similar objects together. Objects within each cluster are similar to each other. The amount of clusters (3) was chosen after applying the “Elbow” method. In this method, sum of squares at each number of clusters (in my case, from 2 to 5) are calculated and plotted. Afterwards, by determining the “elbow” (a point at which the slope changes from steep to shallow), the user can determine an optimal number of clusters. This method is not as precise as other, more mathematically-rigorous cluster-validation methods, but seemed to work for the current (small) dataset.
Indeed, as can be seen from both the matrix and hierarchical clustering, Sehnsucht and Mutter share more similarities with each other than with Rosenrot, which is grouped with Rammstein. (should I include k-means clustering?)
(tba)
Moelants (2002, link tba) showed that there is a certain preferred tempo at around 125 BPM. This natural tempo corresponds with both repeated motor actions and perceived tempo in musical data. As can be seen from the plot, Rammsteins songs mostly fall in 115-140 BPM frame. Interestingly, WEIT WEG does not sound that fast perceptually. This disproportional BPM score might indicate a flaw in the tempo estimation algorithm used by Spotify.
tba
(tbu)
(tbu)
(tbu: playlist, add features that are not 0-1 to a separate table, re-arrange features)
For detailed, low-level audio analysis, typical and atypical songs have to be established. What songs would make you exclaim “yes, of course this is Rammstein”? Personally, I chose Du Hast and Ich will. Of course, this approach is highly subjective and influenced by a plethora of external reasons, such as track popularity, familiarity with the artist, individual biases and tastes. To expand the notion of “typical” to include not only my personal judgements, but also those of a potentially impartial algorithm, standardizes z-scores based on all audio features were computed. The songs that scored nearest to 0, mainly Feuer Frei! and Hallomann were chosen as representatives.
The same approach was taken when selecting atypical songs. In my opinion, Te Quiero Puta! may be the most un-Rammstein song they have ever released. In the context of all their other recordings, this metal-meets-mariachi song sounds novel. Another unique track is Ausländer. It sounds so catchy and pop-y, that it might have been a successful (rock)club song. Unsurprisingly, z-scores indicated that Ein Lied and Roter Sand – Orchester Version differ from the rest of the corpus.
Summary: typical songs - Feuer Frei!, Hallomann, Ich will, Du Hast; atypical songs – Ein Lied, Roter Sand – Orchester Version, Te Quiero Puta!, Ausländer.
| Song | Mode | Key | Tempo | Loudness |
|---|---|---|---|---|
| Du Hast | Minor | A | 125.11 | -6.283 |
| Ich will | Major | D | 128.12 | -4.254 |
| Feuer Frei! | Major | Eb | 95.14 | -3.974 |
| Hallomann | Major | Bb | 172.00 | -5.936 |
| Te Quiero Puta! | Major | F | 159.93 | -5.305 |
| Ausländer | Major | C | 125.03 | -4.159 |
| Ein Lied | Major | F | 122.32 | -19.239 |
| Roter Sand - Orchester Version | Major | G | 82.42 | -10.255 |
1st picture: chordogram for Te Quiero Puta!, 2nd picture: Ausländer (TBU: add titles directly to plots). Althought both of these songs are labeled as atypical by me, they are very different when compared by chordograms. For Te Quiero Puta!, D:maj is noted by a darker band throughout the song. Around 160, a yellow line, indicating a bridge has started. Ausländer chordogram, on the other hand, shows no clear patterns, except for darker bands at F:7 and D:maj.
1st picture: SSM for Ausländer, 2nd picture: SSM for Du hast (TBU: add titles directly to plots). The self-similarity matrix for Ausländer shows very clear repetitiveness in both chroma and timbre plots. The checkerboard pattern shows that a passage is repeated throughout the song. In Du hast timbre plot, we can see a couple of thin (60, 170) yellow lines, which highlight novel parts, whereas in Ausländer, the yellow lines are thicker and re-occur throughout the song.
Rammstein is a German Neue Deutsche Härte band known for heavy riffs, thought-provoking, although controversial, lyrics and flame-fuelled live performances.
Almost unanimously, critics and fans alike agree on the worst Rammstein album, which is said to be Rosenrot (2005). Listeners note that the album is “[…] a disjointed effort glued together with some iron-clad bangers” (Chillingworth, 2019) and “[…] feels like a thrown-together collection of B-sides (because, essentially, it was)” (sean_themighty, 2019).
Interestingly, album rankings oftentimes also agree on the best album – Mutter (2001), describing it as “[…] just….legendary” (JonWood007 , 2020). Furthermore, the album is ranked 324 in Rock Hard magazine’s book of The 500 Greatest Rock & Metal Albums of All Time Rock Hard, 2005. Another strong contender for the title of the most loved Rammstein album, sometimes tied with Mutter, is Sehnsucht (1997). The album contains well-known masterpieces, such as Du hast and Engel.
Additionally, Rammstein has quite a recognizable sound that is usually attributed to distinctive guitars along with Till Lindemann’s baritone and exaggerated trills (so-called rolled r’s).
Thus, in this portfolio, I will investigate whether Spotyfy’s Audio Features and Popularity Scores can provide insights on the following questions:
1. Why Mutter/Sehnsucht are so well-received, whereas Rosenrot is subject to criticism?
2. Rammstein signature sound: what audio features are constant throughout the discography?
3. Typical and atypical Rammstein: what audio analyses can say about Rammstein’s songs?
Corpus
The corpus consists of released studio albums by Rammstein (7 albums, 82 songs), all of which are available on Spotify:
- Herzeleid (1995)
- Sehnsucht (1997)
- Mutter (2001)
- Reise, Reise (2004)
- Rosenrot (2005)
- Liebe ist für alle da (2009)
- Untitled (2019)
Click through the tabs to find out more!
To potentially pinpoint the differences that might have contributed to Mutter/Sehnsucht love and Rosenrot hate, it is useful to take a close-up look at Spotify’s Audio Features. Furthermore, Audio Features were plotted to see which elements change or stay constant throughout the discography.
The first histogram highlights the key distribution per album. Every Rammstein album has notable variation in keys, averaging at 7 keys per album. Mutter seems to be an outlier, with only 4 keys implemented, mainly D, E♭, E and A. Overall, no clear key preference can be seen throughout the discography.
Second plot is a stacked bar chart that illustrates minor/major modes in the corpus. As can be seen, major mode is preferred throughout all albums. This might be a shocker for some, as Rammstein’s music does not sound necessarily positive, which only further demonstrates that major ≠ happy and minor ≠ sad.
On the left, density plots of Spotify Audio Features for Rammstein’s studio albums are presented. Density plots show smoothed distribution of values and the peaks correspond to locations where there is the highest concentration of said values.
Danceability seems to vary per album. It is clear that Sehnsucht (and Herzeleid) have the highest overall danceability. Conversely, the majority of songs from Rosenrot are located lower on the danceability scale.
Interestingly, danceability is one of the most mysterious Spotify’s Audio Features, as the listener’s perception of danceability for a song sometimes conflicts with the value awarded by Spotify. For instance, Du Riechst So Gut has a danceability score of 0.67/1, whereas I’m Looking Forward to Joining You, Finally (Nine Inch Nails) was awarded with a striking score of 0.795/1.
Energy distribution per album shows that overall, Rammstein’s discography boasts high energy. Sehnsucht’s energy is concentrated at around 0.95, while Rosenrot’s energy peak is seen at 0.7.
Next, valence plot marks Rosenrot as a clear outlier. Low valence corresponds to more sad, angry music.
Another interesting visualization is loudness. Both Rosenrot and Sehnsucht exhibit a strong peak at -5. Although Mutter’s highest density shares the same value, the peak is weak.
Finally, regarding tempo, Rosenrot has more songs in 160 (BPM) tempo range than other albums.
Speechiness, liveness, acousticness and instrumentalness do not seem to provide any interesting patterns regarding the three albums in question and also overall.
Conclusion: Mutter, Sehnsucht, Rosenrot are going to be analyzed in-depth based on danceability, energy, valence, loudness and tempo, as these features differ per album.
For those who are unfamiliar with the band, I have compiled a short (10 song) playlist with my favourite songs. Listen to “RAMMSTEIN: Beyond Du Hast” playlist here →
ADD AT THE END
The melody seems to be the strongest in A#/Bb.